DESCRIPTION: The goal of this proposal is to determine the role of two POU domain genes, pdm-1 and pdm-2, in cell differentiation during Drosophila development. POU domain genes encode transcription factors that bind to a consensus octamer sequence and, at least in some cases, their expression has been shown to be linked to cell differentiation. Dr. Sakonju proposes to study the role of the pdm-1 and pdm-2 proteins in neural differentiation in Drosophila. At the onset of neural specification, certain ectodermal cells acquire the potential to follow a neural pathway under the influence of proneural genes. Some of these neural precursor cells become neuroblasts whereas the surrounding cells become epidermal. The pdm-1 and pdm-2 genes are located 50 kb apart in the genome and act in the differentiation pathway of one of these neuroblasts termed NB4-2. This neuroblast divides to give rise to another neuroblast and a ganglion mother cell named GMC-1. The latter then divides to form a terminally differentiated neuron named RP2 and a sib. The pdm-1 and pdm-2 proteins are expressed in GMC-1 but not in RP2/sib. Over-expression of pdm-2 under a heat shock promoter results in the progeny of GMC-1 acquiring a GMC-1 fate, whereas mutations in the pdm-2 gene cause the absence of RP2/sib. This could be because GMC-1 cells do not divide, or because after division the two daughter cells do not differentiate correctly, suggesting in either case that pdm-2 is necessary for correct specification of GMC-1. pdm-1 seems to play a similar and redundant role. The first part of the proposed experiments is aimed at isolating mutations in pdm-1. This will be carried out by several different approaches: 1)screening of existing lethals that fail to complement two large deficiencies of the 33F region; 2) screening for mutations that fail to complement the wing phenotype of nubbin, a wing mutation though to affect the pdm-1 gene; 3) an F2 lethal screen. Once mutations in pdm-1 are obtained, double pdm-1 pdm-2 mutants will be isolated to test whether the two genes have redundant functions. Once mutations in both genes are at hand, their role in the establishement of the NB4-2 lineage will be analyzed. This lineage will first be dissected in detail using two alternative sets of lineage tracer studies, one making clones using the FRT/FLP recombinase system and a second one involving injection of the lipophilic dye DiI. These studies will allow Dr. Sakonju to determine the time at which NB4-2 divides to produce GMC-1 in relationship to the time of pdm expression, the time of GMC-1 division, and the fate of the RP2 sib cell. Once this lineage is unequivocally determined, Dr. Sakonju will study the fate of the GMC-1 cell in pdm-1 and pdm-2 double mutants. If this cell is still present in the mutants, he will determine whether lack of pdms results in the absence of RP2 or incorrect identity of RP2/sib. Once the effect of pdm-1 and pdm-2 on the NB4-2 lineage has been determined, Dr. Sakonju will examine when and where pdm function is required. The time of development when pdm proteins exert their function will be determined by ectopically expressing these proteins at various times during embryogenesis and analyzing the ability to rescue the mutant phenotype. Autonomy of pdm function will be examined by clonal analysis using FRT/FLP. Dr. Sakonju then proposes to analyze which genes activate pdms and which genes are activated by these two proteins by examining regulatory relationships between pdm-1, pdm-2, and candidate genes such as prospero, huckebein, ftz, Ubx, eve, and I-POU. The last part of the proposal is aimed at identifying cis-acting regulatory elements responsible for pdm-2 expression in the NB4-2 lineage. Standard aproaches, fusing various candidate sequences to a lacZ reporter gene and examining their expression after germ line transformation, will be used. Finally, Dr. Sakonju proposes to isolate proteins that interact with pdm-1 and pdm-2 by screening for dominant modifiers of the pdm phenotype.